For older Americans with several chronic conditions, the U.S. health system is "a nightmare to navigate." The quality of care is substandard, and the costs are unsustainable. "Guided Care" is an alternative approach that provides comprehensive, patient-centered, coordinated, evidence-based, IT-enhanced care. In Guided Care, an office-based registered nurse works closely with 2-5 primary care physicians in providing chronic care services to 50-60 older patients who have chronic health conditions. The first 8 months of a 20-month multisite, cluster-randomized controlled trial (cRCT) showed that Guided Care improves the quality of chronic care and tends to reduce its net costs. Unknown, however, are the cumulative effects that Guided Care may produce after the first 20 months, when the nurse-physician teams are clinically mature and the patients'chronic conditions have been well managed. We propose to evaluate the effects of Guided Care, 20-32 months after its introduction, on chronically ill patients'use and cost of health care. We will assess its effects on: 1) patients'use of high-cost services, such as hospital care;2) insurers'net costs for health care provided by hospitals, skilled nursing facilities and home health agencies;and 3) the utilization and costs of care for subgroups of chronically ill older persons, as defined by their morbidity levels and by the characteristics of the delivery systems, primary care practices and insurance plans through which they receive care. Upon completion of the 20-month cRCT of Guided Care (June 2008), a consortium of managed care organizations and a research center provided funding to continue Guided Care for the study participants (n=904) for an additional year (July 2008 - June 2009). During this year, the health services used by the patients in the two groups were documented through insurance claims submitted to Medicare, Kaiser Permanente, or Tricare. We will obtain copies of these claims (covered by existing data use agreements), classify them into nine service categories (e.g., hospital, skilled nursing facility, home health agency), and create a data base containing the service category, date(s), place-of-service, diagnosis and units of service associated with each claim. We will compute the ratio of the two groups'2008-2009 use of each category of service, adjusting for demographic factors, health, function and practice site. Based on these ratios and Medicare's average payments for a unit of each service, we will then compute the difference between the groups'annual health care costs. To compare the two groups'net costs of care, we will also include the costs of providing Guided Care (i.e., the nurses'salaries, benefits, equipment, travel, communication services). Finally, we will repeat these analyses on subgroups of patients to determine which patients and which health care delivery environments benefit the most (and the least) from sustained Guided Care.